• EMP3;
  • methylation;
  • oligodendroglial tumors;
  • overexpression


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results and discussion
  5. References

The epithelial membrane protein 3 (EMP3) gene located on chromosome 19q13 has been implicated as a candidate tumor suppressor gene (TSG) in neuroblastomas and gliomas. The aim of this study was to investigate whether EMP3 is involved in oligodendroglial tumors (OTs), which frequently carry combined chromosomes 1p and 19q deletion. We first investigated the transcript level of EMP3 in a cohort of 57 OTs by quantitative real-time RT-PCR. Our results showed that 10 (18%) tumors had reduced EMP3 expression level compared to normal brains. Six of these tumors carried chromosome 19q13 deletion but no statistical correlation was found between the 2 parameters. Intriguingly, a similar proportion (11 of 57, 19%) of tumors displayed EMP3 overexpression, with 8 of them having transcript level >10-fold higher than normal brain. All 11 OTs retained chromosomes 1p36 and 19q13, and a significant association was found between EMP3 overexpression and balanced chromosomes 1p36 and 19q13 (p = 0.004). The methylation status of EMP3 was evaluated by bisulfite sequencing in 29 OTs with diverse expression levels. All tumors except 3 showed aberrant methylation of EMP3 and no correlation was observed between transcript level and methylation status, suggesting that methylation alone does not mediate transcriptional down-regulation of EMP3 in OTs. In conclusion, our study demonstrates that EMP3 overexpression is involved in OTs retaining chromosomes 1p and 19q and does not support EMP3 as the target TSG on chromosome 19q13 in OTs. © 2006 Wiley-Liss, Inc.

Oligodendroglial tumors (OTs) are primary brain neoplasms that constitute about 5–18% of all gliomas. They comprise the classic oligodendrogliomas and mixed oligoastrocytomas and are tiered into WHO grade II and anaplastic grade III lesions.1 Unlike other glioma subtypes, OTs show distinct clinical features of sensitivity to chemotherapy and longer patient survival.2 The pathogenesis of OTs is poorly understood. A genetic hallmark of these tumors is the combined allelic deletion of chromosomes 1p and 19q, suggesting that these chromosome arms carry critical tumor suppressor genes (TSGs) involved in OT formation.3 These specific genetic alterations have also been demonstrated to be powerful predictors of chemosensitivity and overall survival in OT patients.4

Deletion of chromosome 19q is a common genetic alteration in diffuse gliomas, and is detectable in 71–73% of oligodendrogliomas, 44–68% of oligoastrocytomas and 31–38% of astrocytomas.5, 6 A common deletion region has been mapped to chromosome 19q13.3, and there is an extensive effort in search for the target gene in this region.7 Two genes, PEG3 and p190RhoGAP, have been proposed as candidate TSGs on chromosome 19q13, but their role in gliomagenesis remains elusive.8, 9 Recently, the epithelial membrane protein 3 (EMP3) gene located on chromosome 19q13.32 was shown to have reduced expression in primary neuroblastomas.10 Transcriptional silencing of EMP3 was also detectable in neuroblastoma and glioma cell lines and was associated with aberrant methylation at exon 1 of EMP3. Functional study demonstrated that forced expression of EMP3 in neuroblastoma cells reduced colony formation density and tumor growth in nude mice. Moreover, a screening of 116 primary neuroblastomas and 41 gliomas for aberrant methylation of EMP3 revealed that such epigenetic change was present in 24 and 39% of tumors, respectively. Collectively, these data suggest that EMP3 is a candidate TSG in neuroblastomas and gliomas.10 However, in their study, the histologic subtypes in the glioma series were not specified and it is unclear if EMP3 is the target TSG on chromosome 19q13 in OTs. Hence, the aim of this study was to investigate whether EMP3 was involved in the development of OTs. We evaluated the expression and methylation patterns of EMP3 in a series of 57 OTs.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results and discussion
  5. References

Patients and tumor specimens

A cohort of 57 primary OTs, consisting of 27 oligodendrogliomas (WHO grade II), 14 anaplastic oligodendrogliomas (WHO grade III), 14 oligoastrocytomas (WHO grade II) and 2 anaplastic oligoastrocytomas (WHO grade III), were collected from Prince of Wales Hospital of Hong Kong and Huashan Hospital of Shanghai, China. Tumor tissues resected at the time of surgery were immersed immediately in the RNALater solution (Ambion, TX) and stored at −80°C. All tumors were classified according to current WHO criteria1 and confirmed to have tumor content of at least 80%. The median age of patients was 43 years (range 10–77) and the male/female ratio was 1:1.3. Tumor-matched peripheral blood samples were also collected for 16 patients and served as constitutional controls.

Quantitative real-time reverse transcription-polymerase chain reaction

Total RNA was extracted from tissues with TRIzol reagent and reverse transcribed to cDNA, as previously described.11 Each PCR mix in a 25-μl volume contained 30 ng cDNA, TaqMan Universal PCR Master mix, the commercially available EMP3 or GAPDH probes, and primers (EMP3: Hs00171319_m1 and GAPDH: Hs99999905_m1; Applied Biosystems, CA). Amplification and detection were carried out on the iCycler real-time PCR detection system (Bio-Rad Laboratories, CA) under the conditions of 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min. Standard curves were generated to determine copy numbers of EMP3 and GAPDH. The expression level of EMP3 was normalized against that of reference gene GAPDH for each sample and expressed as fraction of the mean normalized EMP3 level of 8 normal brain tissues. Total RNA of 3 normal brains was obtained from Ambion, Clontech, CA and BioChain, CA.

Bisulfite sequencing

The methylation status of EMP3 was determined by bisulfite sequencing. Genomic DNA extracted from tissues was modified using the Methylamp DNA Modification kit (Epigentek, NY). A 280-bp fragment, which covers 15 CpG sites on exon 1 and intron 1 of EMP3, was amplified from bisulfite-treated DNA with primers 5′-GGGAGTAAGAGAGAAGGAGGT-3′ and 5′-TTAAA AAATCCCAACCCTAAATAAC-3′ at a condition consisting of initial activation at 95°C for 10 min, followed by 35 cycles of 95°C for 30 sec, 62°C for 30 sec and 72°C for 40 sec, and a final extension at 72°C for 10 min. The products were cloned into the pCR2.1-TOPO vector (Invitrogen, CA), and 5–10 positive recombinants were picked and subjected to sequencing as previously described.11 Tumors with at least 50% of clones showing methylation in at least 8 of 15 of CpG sites examined were considered methylation positive.

Allelic deletions on chromosomes 1p36 and 19q13

The allelic statuses of chromosomes 1p36 and 19q13 in the tumor series were determined by either microsatellite analysis (in samples with matched peripheral blood) or by fluorescent in situ hybridization (FISH) as reported.12 The polymorphic loci examined by microsatellite analysis were D1S468, D1S508, D1S1612, D1S199, D1S2743, D19S219, D19S112 and D19S412. The target BAC probes for chromosomes 1p36.3 and 19q13.3 were RP11-62M23 and CTD-2571L23, respectively; and the reference BAC probes for chromosomes 1q31 and 19p12 were RP11-162L13 and RP11-420K14, respectively.

Statistical analysis

Statistical analysis was performed using the software SPSS 10.0. The correlation between the 2 parameters was evaluated by Chi-square test or Fisher's exact test, whichever was appropriate. An obtained p value less than 0.05 (2-sided) was considered statistically significant.

Results and discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results and discussion
  5. References

To evaluate whether EMP3 was involved in OTs, we first quantified its transcript level in a series of 57 OTs. Our quantitative real-time reverse transcription-polymerase chain reaction (Q-RT-PCR) analysis demonstrated that 10 (18%) tumors had reduced EMP3 expression level (0.1–0.3) compared to 8 normal brains (Table I). Six of these tumors also carried chromosome 19q13 deletion but no statistical correlation (p = 0.308, Fisher's Exact test) was found between reduced EMP3 expression and chromosome 19q13 loss (Table II). Decreased EMP3 expression was however statistically associated with chromosome 1p36 deletion (p = 0.035). Intriguingly, 11 (19%) tumors displayed enhanced EMP3 expression, with 8 of them having transcript level 10- to 50-fold higher than normal brains. All of these 11 OTs were allelic balanced at both chromosomes 1p36 and 19q13 (Table II). Overexpression of EMP3 was significantly associated with balanced 19q13 (p = 0.002), balanced 1p36 (p = 0.001) and balanced 1p36/19q13 (p = 0.004). No statistical correlations were found between EMP3 expression level (up- or down-regulated) and histologic subtype, tumor grade, age or gender.

Table I. EMP3 Expression in 57 OTs Examined
 Relative EMP3 expression1Allelic loss
Down-regulationUp-regulationChr 1pChr 19q
  • Values inside parentheses indicate percentages.

  • 1

    Compared with the mean EMP3 expression level of 8 normal brain tissues.

Oligodendroglioma (n = 27)6 (22)03 (11)15 (56)13 (48)
Anaplastic oligodendroglioma (n = 14)2 (14)2 (14)1 (7)5 (36)5 (36)
Oligoastrocytoma (n = 14)2 (14)1 (7)2 (14)7 (50)6 (43)
Anaplastic oligoastrocytoma (n = 2)002 (100)00
Total (n = 57)10 (18)3 (5)8 (14)27 (47)24 (42)
Table II. Allelic Status of Chromosome 1p and 19q in OTs with Aberrant EMP3 Expression
 Tumors with down-regulated EMP3 expressionTumors with up-regulated EMP3 expression
Allelic loss of chr 1pAllelic loss of chr 19qAllelic loss of chr 1pAllelic lossof chr 19q
  • Values inside parentheses indicate n values.

  • *

    Correlation between EMP3 expression and allelic status of chromosome 1p or 19q by Fisher's Exact test.

Oligodendroglioma5 (6)40 (3)0
Anaplastic oligodendroglioma2 (2)10 (3)0
Oligoastrocytoma1 (2)10 (3)0
Anaplastic oligoastrocytoma0 (0)00 (2)0

Since transcriptional silencing of EMP3 has been linked with aberrant methylation in neuroblastoma and glioma cell lines,10 we then investigated the potential involvement of such epigenetic alteration in tumors with decreased EMP3 transcript level. Two CpG islands are mapped on EMP3, one locates to the noncoding exon 1 and part of intron 1 and the other to the putative promoter region. We determined the methylation status of 15 CpG sites located at exon 1 and intron 1 of EMP3 in 29 OTs, for which genomic DNA was available, and 3 normal brains by bisulfite sequencing. Our results showed that there was no methylation or only limited methylation in a few CpG sites at exon 1 and intron 1 of EMP3 in 3 normal brains examined (Fig. 1). In contrast, aberrant methylation was found in all or majority of the 15 CpG sites in tumors with reduced MP3 expression. Unexpectedly, we also detected dense methylation in tumors, except HS364, HS390 and HS395, with normal and elevated EMP3 expression levels. There was no statistical correlation between the methylation status of EMP3 and its expression level or chromosome 19q13 deletion. We further examined the methylation status of another CpG island located at the promoter region of EMP3 in OTs with diverse EMP3 expression levels and in normal brains. All CpG sites studied were methylated in both normal and tumor samples (data not shown). Taken together, these results indicate that aberrant methylation does not mediate transcriptional down-regulation of EMP3 in OTs.

thumbnail image

Figure 1. Summary of bisulfite sequencing analysis of EMP3 in 29 OTs and 3 normal brains (NB) studied. (a) Schematic representation of 15 CpG sites at exon 1 (shaded box) and intron 1 of EMP3. Each vertical line represents a CpG site. Arrows indicate forward and reverse primers used in bisulfite sequencing. (b) The methylation status of EMP3 is indicated by circles, with each circle representing the average methylation level at a CpG site determined from 5–10 clones. Open cicle, 0% methylation; quarter-filled circle, 1–25% methylation; half-filled circle, 26–50% methylation; three-quarter-filled circle, 51–75% methylation; filled circle, >75% methylation. EMP3 expression in tumors is expressed as a fraction of the mean EMP3 transcript level of 8 normal brain tissues. The allelic statuses of chromosomes 1p and 19q are determined either by FISH or microsatellite analysis. ND, not determined.

Download figure to PowerPoint

EMP3 belongs to the PMP22 gene family which includes PMP22, EMP1, EMP2 and the more distant member MP20. All family members encode proteins with 4 predicted transmembrane domains and N-glycosylation sites on the extracellular loop [reviewed in Ref.13]. PMP22 is the best characterized family member. Its product is a component of the compact peripheral nerve myelin and is involved in several human hereditary peripheral neuropathies. Notably, it has been demonstrated that PMP22 is expressed at high levels in growth arrested NIH-3T3 fibroblasts, and forced expression of this gene induced an apoptotic-like phenotype in these cells.14 On contrary, enhanced expression of EMP1 was observed in NIH-3T3 fibroblasts during cell proliferation.15 These results suggest that PMP22 and EMP1 may play opposite regulatory roles in cell growth. The biological function of EMP3 remains unclear.

In this study, we presented data that argue against EMP3 as the target TSG on chromosome 19q13 in OTs. First, there were similar proportions (∼18%) of OTs showing either up- or down-regulated EMP3 expression. Notably, a majority (8 of 11) of tumors with EMP3 overexpression had markedly elevated transcript levels at least 10-fold higher than normal. Second, if EMP3 were the target gene on chromosome 19q13, we speculated that allelic deletion on chromosome 19q13 would be a major mechanism contributing to down-regulation of EMP3. However, only 6 of 10 tumors with reduced EMP3 transcript levels were detected with such chromosomal aberrations, and there was no correlation between reduced EMP3 expression and chromosome 19q13 loss status. In contrast, all tumors with EMP3 overexpression were demonstrated to have balanced chromosomes 1p36 and 19q13 and the correlation was statistically significant. These data suggest that EMP3 upregulation is involved in those OTs without the combined chromosomal deletion. Moreover, aberration methylation of EMP3 was found in all but 3 OTs examined and was not correlated with EMP3 expression. Our results indicate that EMP3 transcript level is independent of methylation in OTs.

The finding of enhanced EMP3 expression in OTs with balanced chromosomes 1p36 and 19q13 suggests that EMP3 profile may differentiate distinct subgroups of OTs. It has been well documented that OTs with combined 1p/19q deletion are closely associated with better overall patient survival and responsiveness to chemotherapy.4, 16 Whether the EMP3-overexpressing OTs behave clinically different from OTs with 1p/19q deletion, for example their prognosis and chemosensitivity, requires further investigation.

While this article was undergoing revision, Tews et al. reported the expression profiles of all genes residing on 1p36.13-p36.31 and 19q13.2-q13.33 in a series of 27 OTs and 8 diffuse astrocytomas using cDNA microarray.17 EMP3 was found to be one of the most differentially expressed genes in 1p/19q-intact tumors compared with 1p/19q-deleted tumors. However, statistical analysis failed to demonstrate that the differential EMP3 expression was significant between these tumor groups. Moreover, although our finding of differential EMP3 expression profile in OTs was in line with theirs, we were unable to draw further conclusions on the 2 data sets due to the following reasons: sensitivity of techniques used (real-time RT-PCR vs. microarray), inclusion of nonOTs in their series and the different reference RNA sources used in expression analysis (normal brain vs. commercially available universal reference RNA).

In conclusion, our study demonstrates that enhanced EMP3 expression is associated with OTs retaining both chromosomes 1p36 and 19q13 and does not support EMP3 as the target TSG on chromosome 19q13 in OTs.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results and discussion
  5. References
  • 1
    Reifenberger G,Kros JM,Burger PC,Louis DN,Collins VP. Oligodendroglial tumors. In: KleihuesP,CaveneeWK, eds. Pathology and genetics—tumours of the nervous system. Lyon: IARC Press, 2000; 5669.
  • 2
    Perry JR,Louis DN,Cairncross JG. Current treatment of oligodendrogliomas. Arch Neurol 1999; 56: 4346.
  • 3
    Bello MJ,Leone PE,Vaquero J,de Campos JM,Kusak ME,Sarasa JL,Pestana A,Rey JA. Allelic loss at 1p and 19q frequently occurs in association and may represent early oncogenic events in oligodendroglial tumors. Int J Cancer 1995; 64: 20710.
  • 4
    Cairncross JG,Ueki K,Zlatescu MC,Lisle DK,Finkelstein DM,Hammond RR,Silver JS,Stark PC,Macdonald DR,Ino Y,Ramsay DA,Louis DN. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998; 90: 14739.
  • 5
    Smith JS,Alderete B,Minn Y,Borell TJ,Perry A,Mohapatra G,Hosek SM,Kimmel D,O'Fallon J,Yates A,Feuerstein BG,Burger PC, et al. Localization of common deletion regions on 1p and 19q in human gliomas and their association with histological subtype. Oncogene 1999; 18: 414452.
  • 6
    Von Deimling A,Fimmers R,Schmidt MC,Bender B,Fassbender F,Nagel J,Jahnke R,Kaskel P,Duerr E,Koopmann J,Maintz D,Steinbeck S, et al. Comprehensive allelotype and genetic analysis of 466 human nervous system tumors. J Neuropathol Exp Neurol 2000; 59: 54458.
  • 7
    Hartmann C,Johnk L,Kitange G,Wu Y,Ashworth LK,Jenkins RB,Louis DN. Transcript map of the 3.7-Mb D19S112–D19S246 candidate tumor suppressor region on the long arm of chromosome 19. Cancer Res 2002; 62: 41008.
  • 8
    Maegawa S,Yoshioka H,Itaba N,Kubota N,Nishihara S,Shirayoshi Y,Nanba E,Oshimura M. Epigenetic silencing of PEG3 gene expression in human glioma cell lines. Mol Carcinog 2001; 31: 19.
  • 9
    Wolf RM,Draghi N,Liang X,Dai C,Uhrbom L,Eklof C,Westermark B,Holland EC,Resh MD. p190RhoGAP can act to inhibit PDGF-induced gliomas in mice: a putative tumor suppressor encoded on human chromosome 19q13.3. Genes Dev 2003; 17: 47687.
  • 10
    Alaminos M,Davalos V,Ropero S,Setien F,Paz MF,Herranz M,Fraga MF,Mora J,Cheung NK,Gerald WL,Esteller M. EMP3, a myelin-related gene located in the critical 19q13.3 region, is epigenetically silenced and exhibits features of a candidate tumor suppressor in glioma and neuroblastoma. Cancer Res 2005; 65: 256571.
  • 11
    Chang Q,Pang JC,Li KK,Poon WS,Zhou L,Ng HK. Promoter hypermethylation profile of RASSF1A, FHIT, and sFRP1 in intracranial primitive neuroectodermal tumors. Hum Pathol 2005; 36: 126572.
  • 12
    Dong Z,Pang JC,Ng MH,Poon WS,Zhou L,Ng HK. Identification of two contiguous minimally deleted regions on chromosome 1p36.31–p36.32 in oligodendroglial tumours. Br J Cancer 2004; 91: 110511.
  • 13
    Jetten Am,Suter U. The peripheral myelin protein 22 and epithelial membrane protein family. Prog Nucleic Acid Res Mol Biol 2000; 64: 97129.
  • 14
    Fabbretti E,Edomi P,Brancolini C,Schneider C. Apoptotic phenotype induced by overexpression of wild-type gas3/PMP22: its relation to the demyelinating peripheral neuropathy CMT1A. Gene Dev 1995; 9: 184656.
  • 15
    Ben-Porath I,Benvenisty N. Characterization of a tumor-associated gene, a member of a novel family of genes encoding membrane glycoproteins. Gene 1996; 183: 6975.
  • 16
    van den Bent MJ,Looijenga LHJ,Langenberg K,Dinjens W,Graveland W,Uytdewilligen L,Smitt PAS,Jenkins RB,Kros JM. Chromosomal anomalies in oligodendroglial tumors are correlated with clinical features. Cancer 2003; 97: 127684.
  • 17
    Tews B,Felsberg J,Hartmann C,Kunitz A,Hahn M,Toedt G,Neben K,Hummerich L,von Deimling A,Reifenberger G,Lichter P. Identification of novel oligodendroglioma-associated candidate tumor suppressor genes in 1p36 and 19q13 using microarray-based expression profiling. Int J Cancer 2006; 119: 792800.